Linux NFS development
 help / color / mirror / Atom feed
From: NeilBrown <neilb@ownmail.net>
To: Alexander Viro <viro@zeniv.linux.org.uk>,
	Christian Brauner <brauner@kernel.org>,
	Chuck Lever <cel@kernel.org>, Jeff Layton <jlayton@kernel.org>,
	Jori Koolstra <jkoolstra@xs4all.nl>,
	Mateusz Guzik <mjguzik@gmail.com>,
	Dorjoy Chowdhury <dorjoychy111@gmail.com>
Cc: Trond Myklebust <trondmy@kernel.org>,
	Anna Schumaker <anna@kernel.org>,
	Andreas Gruenbacher <agruenba@redhat.com>,
	gfs2@lists.linux.dev, Ilya Dryomov <idryomov@gmail.com>,
	Alex Markuze <amarkuze@redhat.com>,
	Viacheslav Dubeyko <slava@dubeyko.com>,
	ceph-devel@vger.kernel.org, Paulo Alcantara <pc@manguebit.org>,
	Namjae Jeon <linkinjeon@kernel.org>,
	linux-cifs@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	linux-nfs@vger.kernel.org
Subject: [PATCH v2 08/14] vfs: don't return -ENODEV from vfs_lookup_open()
Date: Sat, 19 Sep 2026 12:06:12 +1000	[thread overview]
Message-ID: <20260919022441.3305170-9-neilb@ownmail.net> (raw)
In-Reply-To: <20260919022441.3305170-1-neilb@ownmail.net>

From: NeilBrown <neil@brown.name>

-ENODEV is not generally an error meaning "the object is a device"
and nfsd - the only intended caller of vfs_lookup_open() - does not
benefit from knowing it was a device file.  So return -EFTYPE in
that case.
Also switch to testing the dentry type rather than dereferencing the
inode to get the type.

-EISDIR is widely used to mean "the object is a directory which is
   not what is wanted".
-ELOOP is sometimes used elsewhere to mean "a symlink was found but
   cannot be handled".

Also remove note about ->atomic_open returning -EFTYPE as that now only
happens for non regular/dir/symlink.

Signed-off-by: NeilBrown <neil@brown.name>
---
 fs/namei.c | 24 +++++-------------------
 1 file changed, 5 insertions(+), 19 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index b9fca38ad489..ba3e7e4b5fdb 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -4621,9 +4621,7 @@ out_dput:
  * determine the type of file found from the error.
  * -EISDIR : a directory was found
  * -ELOOP  : a symlink was found
- * -ENODEV : a block or character device special file was found
- * -EFTYPE : any other non-regular file was found, such as FIFO or SOCK.
- *           or ->atomic_open responded to __O_REGULAR.
+ * -EFTYPE : any other non-regular file was found, device-special, FIFO or SOCK
  *
  * Returns: the opened struct file, or an error.
  */
@@ -4671,24 +4669,12 @@ struct file *vfs_lookup_open(struct path *parent, struct qstr *last,
 		error = -ENOENT;
 	} else if (!(file->f_mode & FMODE_CREATED) && (open_flag & O_EXCL)) {
 		error = -EEXIST;
-	} else if ((dentry->d_inode->i_mode & S_IFMT) != S_IFREG) {
-		switch (dentry->d_inode->i_mode & S_IFMT) {
-		case S_IFDIR:
+	} else if (!d_is_reg(dentry)) {
+		error = -EFTYPE;
+		if (d_is_dir(dentry))
 			error = -EISDIR;
-			break;
-		case S_IFLNK:
+		if (d_is_symlink(dentry))
 			error = -ELOOP;
-			break;
-		case S_IFBLK:
-		case S_IFCHR:
-			error = -ENODEV;
-			break;
-		case S_IFIFO:
-		case S_IFSOCK:
-		default:
-			error = -EFTYPE;
-			break;
-		}
 	} else if (!(file->f_mode & FMODE_OPENED)) {
 		nd.path.dentry = dentry;
 		error = vfs_open(&nd.path, file);
-- 
2.50.0.107.gf914562f5916.dirty


  parent reply	other threads:[~2026-09-19  2:26 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-19  2:06 [PATCH v2 01/14] fixes for vfs_lookup_open, and integration with nfsd NeilBrown
2026-09-19  2:06 ` [PATCH v2 01/14] VFS: revise and expand documentation for atomic_open NeilBrown
2026-09-24 12:25   ` Jeff Layton
2026-09-19  2:06 ` [PATCH v2 02/14] nfs: correctly handle NFS4ERR_WRONG_TYPE from v4 OPEN request NeilBrown
2026-09-24 12:55   ` Jeff Layton
2026-09-24 13:01   ` Jeff Layton
2026-09-19  2:06 ` [PATCH v2 03/14] gfs2: simplify atomic_open handling NeilBrown
2026-09-19 16:49   ` Andreas Gruenbacher
2026-09-19 22:22     ` NeilBrown
2026-09-20 16:31       ` Andreas Gruenbacher
2026-09-22 21:16         ` NeilBrown
2026-09-19  2:06 ` [PATCH v2 04/14] ceph: simplify atomic_open to use finish_no_open() NeilBrown
2026-09-24 13:02   ` Jeff Layton
2026-09-25 21:49     ` NeilBrown
2026-09-26 11:48       ` Jeff Layton
2026-09-19  2:06 ` [PATCH v2 05/14] cifs: allow -EISDIR precedence over -EFTYPE in __cifs_do_create() NeilBrown
2026-09-23  5:54   ` Namjae Jeon
2026-09-23  6:50     ` NeilBrown
2026-09-23  7:52       ` Namjae Jeon
2026-09-23  8:13   ` Namjae Jeon
2026-09-19  2:06 ` [PATCH v2 06/14] vfs: add some allowed open flags to vfs_lookup_open() NeilBrown
2026-09-24 12:54   ` Jeff Layton
2026-09-29 15:18   ` Jori Koolstra
2026-09-29 22:04     ` NeilBrown
2026-09-19  2:06 ` [PATCH v2 07/14] vfs: O_NONBLOCK|O_CREAT open shouldn't wait for directory delegation NeilBrown
2026-09-24 12:51   ` Jeff Layton
2026-09-19  2:06 ` NeilBrown [this message]
2026-09-24 13:07   ` [PATCH v2 08/14] vfs: don't return -ENODEV from vfs_lookup_open() Jeff Layton
2026-09-19  2:06 ` [PATCH v2 09/14] vfs: change vfs_lookup_open() to use do_open(), not vfs_open() NeilBrown
2026-09-24 13:13   ` Jeff Layton
2026-09-19  2:06 ` [PATCH v2 10/14] nfsd: make EEXIST checks in nfsd4_create_file() more consistent NeilBrown
2026-09-24 13:23   ` Jeff Layton
2026-09-25 21:57     ` NeilBrown
2026-09-19  2:06 ` [PATCH v2 11/14] nfsd: check for mountpoints after non-creating open NeilBrown
2026-09-24 13:03   ` Jeff Layton
2026-09-25 22:14     ` NeilBrown
2026-09-19  2:06 ` [PATCH v2 12/14] nfsd: switch NFS4 OPEN to use vfs_lookup_open() NeilBrown
2026-09-20 17:10   ` Chuck Lever
2026-09-22 21:55     ` NeilBrown
2026-09-23  4:07       ` NeilBrown
2026-09-19  2:06 ` [PATCH v2 13/14] nfsd: change nfsd_check_obj_isreg() to use nfs error codes NeilBrown
2026-09-24 13:25   ` Jeff Layton
2026-09-19  2:06 ` [PATCH v2 14/14] nfsd: use vfs_lookup_open() for non-creating open requests too NeilBrown
2026-09-20 17:13   ` Chuck Lever
2026-09-25 22:26     ` NeilBrown
2026-09-25 16:04 ` [PATCH v2 01/14] fixes for vfs_lookup_open, and integration with nfsd Christian Brauner
2026-09-25 16:56   ` Chuck Lever

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260919022441.3305170-9-neilb@ownmail.net \
    --to=neilb@ownmail.net \
    --cc=agruenba@redhat.com \
    --cc=amarkuze@redhat.com \
    --cc=anna@kernel.org \
    --cc=brauner@kernel.org \
    --cc=cel@kernel.org \
    --cc=ceph-devel@vger.kernel.org \
    --cc=dorjoychy111@gmail.com \
    --cc=gfs2@lists.linux.dev \
    --cc=idryomov@gmail.com \
    --cc=jkoolstra@xs4all.nl \
    --cc=jlayton@kernel.org \
    --cc=linkinjeon@kernel.org \
    --cc=linux-cifs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=mjguzik@gmail.com \
    --cc=neil@brown.name \
    --cc=pc@manguebit.org \
    --cc=slava@dubeyko.com \
    --cc=trondmy@kernel.org \
    --cc=viro@zeniv.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox